Abstract

Diabetes mellitus (DM) and hypertension (HT) are known risk factors for myocardial dysfunction, up till now main focus was on systolic dysfunction but subtle diastolic relaxation abnormalities which appears earlier is overlooked. In our study we compared left ventricular (LV) function by doppler imaging in patients with diabetes mellitus with and without systemic hypertension. We divided patients in four groups: HT, DM, HT + DM and controls. Age and ejection fraction were comparable in all groups and we concluded that diastolic dysfunction was detected with conventional doppler in HT and hypertension + diabetes. But conventional doppler was unable to detect diastolic dysfunction in DM while tissue doppler imaging was able to detect diastolic dysfunction in all groups hence we concluded tissue doppler imaging is better modality for detection of early diastolic abnormalities.

Keywords: Diabetes mellitus, hypertension, myocardial dysfunction, diastolic dysfunction  

Diabetes mellitus (DM) and hypertension (HT) are important risk factors for the development of coronary heart disease. Till date, all the emphasis is focused on systolic function and its implication as clinical outcome. Many patients with DM with or without HT have normal left ventricular (LV) systolic function but still they complains of dyspnea, peripheral edema, raised jugular venous pressure (JVP) and other sign and symptoms of congestive heart failure. Therefore there is need for assessment of diastolic dysfunction in diabetes and HT with normal LV systolic function. 

DM is an established risk factor for cardiovascular events especially the development of congestive heart failure (Kannel WB, et al).¹ It has been suggested that impairment of LV function in patients with DM is due to concomitant risk factors such as arterial HT or diffuse peripheral and coronary atherosclerosis (Ruderman NB, et al).² However, actual mechanisms remain unclear and evidence has accumulated for the existence of a distinct diabetic cardiomyopthy.

Impairment of LV diastolic function in patients with DM has been described using digitized M-mode and doppler echocardiography. Some of the previous studies had proved association of diabetes and diastolic dysfunction while some have proved no association between DM and diastolic dysfunction. Most of the previous studies are done by using conventional doppler for estimation of diastolic dysfunction. Therefore, in our study we have used in addition to conventional doppler echocardiography, tissue doppler imaging for assessment of LV function – A new technique that is independent of loading conditions.

Our study demonstrated the early appearance of both LV systolic and diastolic dysfunction in diabetes patients at rest and the contributory effect of diabetes to myocardial impairment produced by HT, as well as the high usefulness of tissue doppler imaging in detection and quantitation of myocardial dysfunction in diabetics. This method was superior to other echocardiographic techniques. 

Material and Methods

The study was conducted on patients who presented to the outpatient department or were admitted as in patient either in Cardiology ward or the Coronary Care Unit of Sir Sunderlal Hospital, BHU, Varanasi between February 2008 to July 2009. Three types of patient were included in study: Hypertensive 1st time diagnosed not under treatment, diabetic, hypertensive + diabetic.

Age and sex matched control patients were also taken. A detail clinical history was taken and physical examination was performed. TMT (Treadmill Test) was done in all patients and only those patients whose TMT was negative for inducible myocardial ischemia were registered in study. History of ischemic chest pain, poor quality echocardiographic imaging, valvular or congenital heart disease, decreased LV ejection fraction i.e. <55%, endocrine disorder other than diabetes, absence of stable sinus rhythm, renal impairment with serum creatinine >2.5 mg%, conductive or rhythm disturbances on ECG were excluded from the study.

All study participants underwent standard echo-cardiography with doppler studies using Seimen’s Acuson CV70 machine with a 2.5 MHz transducer. The measurements of LV dimension and wall thickness were performed from 2-dimensionally targeted M-mode tracings. LV mass was calculated by standard formula and indexed for body surface area to obtain the LV mass index. LV ejection fraction was estimated using Simpson’s biplane method and LV Teich (M) method. LV fractional shortening was derived from standard formula.

Conventional doppler measurements was done with the transducer in the apical position the sample value is placed at the tip of the mitral valve leaflet and by using pulse wave doppler the inflow pattern of mitral valve is evaluated. The early flow coincident with mitral E wave peak velocity was noted. Along with E wave second wave A wave peak velocity was noted and ratio of E/A is calculated.

Calculation of Deceleration Time

Deceleration time (DT) is commonly used parameter for assessment of diastolic dysfunction. The slope of descending limb of E wave is measured and deceleration time is calculated. Measurement of deceleration time of E wave is problematic when E wave decay curve is not linear, in these instances, using the mid portion of E wave deceleration is the most appropriate for determining slope. Many individuals have a Ski slope E wave in which there is a short rapid decline is E wave velocity followed by a more gradual slope. In this instance ignoring the initial steep decline and measuring the mid portion of the E wave is the appropriate methodology. In many instances the E wave velocity does not decline fully to a velocity of 0 m/s and hence the true time points for DT cannot be determined, standard practice is to extrapolate the DT from the more proximal portions of the E wave.

Tissue Doppler Imaging

Tissue doppler imaging was performed in apical views 4 and 2 chambers and in parasternal short axis view to assess the function of longitudinal and circumferential myocardial fibers. The sampling window was positioned as parallel as possible with the myocardial segment of interest to ensure the optimal angle of imaging the sample values were located in the central part of the basal and mid segments in each apical view and mid anteroseptal and posterior segments in the parasternal short axis view.

Mean values of peak systolic velocities and peak early and peak late diastolic velocities obtained form the basal segments served as a measure of global myocardial function.

For the mid and basal segments, we estimated the mean peak systolic velocity, mean peak early/peak late diastolic velocity by keeping the sample values in central part of mid and basal segments.

Data were analyzed using standard statistical software SPSS (version 12.0). The statistical analysis was done using Analysis of variance (ANOVA) and for group comparison Bonferroni test was used. Chi-square test was used for analyzing qualitative data. P <0.05 considered as statistically significant.

Results

In group I (Hypertensive group) there was 20 patients out of which 7 (35%) were females. Mean age in group I was 55.40 ± 4.59 years. In group II only diabetic (DM) patients were included with mean age was 57.60 ± 8.15 years and there was 10 patients in DM group with only 2 (20%) female patients. In group III (n = 20) both diabetic and hypertensive patients were included
with mean age 55.35 ± 6.74 years. There was 7 (35%) female patients in this group. In group IV 20 control
with 4 female (20.0%) was included with mean age of 56.20 ± 4.37 years.

In group I the mean systolic blood pressure (SBP) of 161.0 ± 1 1.26 mmHg and mean diastolic BP (DBP) 100.40 ± 7.38 mmHg. In group II mean SBP of 120.20 ± 8.76 mmHg and mean DBP of 77.80 ± 7.80 mmHg. In group III mean SBP was 153.30 ± 10.34 mmHg and mean DBP was 97.80 ± 8.60 mmHg. Mean SBP in control group was 122.27 ± 8.31 mmHg and mean DBP was 79.60 ± 6.64 mmHg. On statistical analysis age in all 4 groups was comparable.

Mean LV ejection fraction in HT group was 66.77 ± 3.66%, mean LV ejection fraction in DM group was 64.00 ± 3.52%, mean LV ejection fraction in diabetic and HT group was 65.15 ± 4.17% and mean LV ejection fraction in control group was 63.93 ± 4.38%. On comparing each group with control and group I with group III and group II and group III. P-value was not statistically significant indicating there was no impact of HT and DM on ejection fraction.

In HT group LV mass index was 145.5 ± 8.7 gm/m² while in control group it was 98.60 ± 8.63 gm/m², p-value was <0.001. In DM group LV mass index was 95.50 ± 7.51 gm/m² (p = NS). In HT + DM group mean LV mass index was 151.35 ± 8.13 gm/m² (p <0.001). On comparison with control there was significant increase in LV mass index in HT group greater increase was in HT + DM group. On comparing diabetic with HT + DM group (p <0.001) concluding that DM by it self does not increase LV mass index. HT increases LV mass index and greatest increase was in HT + DM group (Table 1).

In HT group E/A was 0.82 ± 0.22 while in control group E/A was 1.15 ± 0.15 (p <0.001). In DM group E/A ratio was 1.12 ± 0.21 comparing with control (p = NS). In HT + DM group E/A ratio was 0.76 ± 0.15 comparing with control (p <0.001). On comparing HT with HT + DM group (p = NS). HT was associated with decrease in E/A ratio more pronounced in HT with DM. But independently only DM was not associated with significant change in E/A ratio (Table 2). 

Mean DT in HT group was 209.00 ± 23.07 msec while in control group it was 173.47 ± 24.34 m/sec on comparison (p <0.001). Mean DT in DM group was 200.00 ± 18.8 m/sec as comparing with control (p = 0.05). Mean DT in HT + DM group was 233.20 ± 24.23 m/sec on comparing with control (p <0.001). On comparing DM with HT + DM group p-value was significant (p = 0.003). On comparing HT with DM + HT group (p <0.001). DT is increased in HT. It is not significantly increased in diabetic patient and in cases with both HT + DM group DT is increased to greatest extend (Table 3). 

MPSV (mean peak systolic velocity) in HT group was 6.33 ± 0.93 cm/s while in control group it was 7.62 ± 0.74 (p <0.001). MPSV in DM group was again 6.33 ± 0.42 both of these were significantly less than control group (p <0.001). In HT + DM group velocity was least 5.79 ± 0.49 and was again significantly less than control group (p <0.001). While comparing DM with HT + DM group and HT with DM + HT group result was insignificant. Mean positive systolic velocity in basal segments was significantly decreased in patients with diabetic, HT and DM + HT (Table 4). 

E’ (mean peak early myocardial diastolic velocity) in HT group was 6.14 ± 0.82 cm/sec as comparing with control in which it was 7.64 ± 1.01 cm/sec, it was significantly decreased (p <0.001). E’ in DM group was 6.63 ± 0.38 cm/sec as comparing with control it was also decreased (p <0.001). E’ in HT + DM group was 4.79 ± 0.61 cm/sec which was least in all study groups and was significantly less than control (p <0.001). In both HT and DM group E’ was significantly less than control while E’ was least in HT + DM group.

In HT group E’/A’ was 0.90 ± 0.18 while in control group E’/A’ was 1.21 ± 0.16, on comparison E’/A’ was significantly decreased in HT group versus control group (p <0.001). In diabetic group E’/A’ was 1.11 ± 0.21 which was not significantly different than control group (p = 1.000). In HT + DM group E’/A’ was 0.76 ± 0.11 which was very significantly less control group (p <0.001). On comparing DM group with HT + DM group, E’/A’ was significantly less in combination group (HT + DM) (p <0.001). While E’/A’ was not significantly different in HT and HT + DM group (p = 0.06) (Table 5). 

MPSV (mid segment) in HT group was 4.72 ± 0.86 cm/sec which was significantly less than in control group in which it was 6.44 ± 0.42 cm/sec (p <0.001). MPSV in DM group was 4.74 ± 0.48 which was significantly less than control group (p <0.001). MPSV in HT + DM group was 4.49 ± 0.39 cm/sec, which was least in all groups and was significantly less than control group (p <0.001). While MPSV in HT + DM group was not significantly different than HT or DM group.

E’ (mid segment) in HT group was 4.85 ± 0.67 cm/sec which was significantly less in control group in which it was 6.30 ± 0.41 cm/sec (p <0.001). E’ (mid) in DM group was 5.22 ± 0.58 cm/sec, which was significantly less than control group (p <0.001). E’ (mid) in HT + DM group was 3.77 ± 0.55 cm/sec, which was significantly less than control group (p = 0.001) on comparing HT + DM group with DM or HT velocities were significantly less in HT + DM group (p <0.001).

E’/A’ (mid segment) was 0.89 ± 0.19 in HT group which was significantly less than control group in which it was 1.15 ± 0.15 (p <0.001). E’/A’ (mid) was 1.09 ± 0.20 in DM group which was not significantly different than control group (p = 1.000). E’/A’ in HT + DM group was 0.77 ± 0.11 which was significantly less than control group (p <0.001). E’/A’ in HT + DM group was significantly less than DM group (p <0.001) while E’/A’ was insignificantly different in HT and HT + DM group (p = 0.15).

Discussion

The present study represents assessment of LV diastolic filling in adults having DM with or without HT in a population based sample of middle aged and older adults. The present study reveals association between DM and diastolic filling that is independent of age, BP and systolic function. The filling pattern in hypertensive is compared to control and impact of HT plus DM as LV filling pattern is also studied.

In our study LV mass index in DM group (95.50 ± 7.5 gm/m²) was not significantly different than control (98.60 ± 8.63 gm/m²). Kosmala W et al³ in their study showed similar finding. In HT group LV mass index was 145.55 ± 8.73 gm/m² and significantly more than control and was greatest in HT + DM group. Kucharski W et al³ in their study showed similar findings.

Liu JE⁴ in 2001 showed similar findings that LV mass index was indifferent in control versus DM group and greatest increase was in HT + DM group while in HT group LV mass index was increased to moderate range.

In our study E/A ratio was comparable in DM and control group. E/A ratio was less in HT group and was least in HT + DM group. Kosmala W et al³ in their study showed similar trend.

DT was insignificantly different between control and DM group and DT was increased significantly in HT group and was maximal in HT + DM group. Kucharski W et al³ in their study found similar comparable findings. Liu JE et al⁴ in their strong heart study showed similar findings with HT group and HT + DM group but they showed by DT and E/A ratio significant impairment in LV relaxation in DM group their findings might be different in DM group from our study as they included patients mainly with poor glycemic control and with microvascular complication. 

Shapiro LM et al⁵ proved the association of diastolic filling abnormalities with DM, it was also concluded by them that the abnormality of relaxation was more severe in the combined DM + HT group, suggesting additive deleterious effect on active LV relaxation in early diastole when both these conditions are present.

MPSV was comparable in HT and DM group but was significantly less than control group. MPSV was least in HT + DM combined group indicating additive effect of each on impaired ventricular relaxation.

Kosmala W³ in their study in 2004 also found that longitudinal systolic myocardial function evidenced by decreased mean peak systolic velocity was reduced in all the patient groups being most severely effected in the DM + HT group.

Velocities in the mid segments were again reduced in DM group and HT group as compared with control and velocities were least in HT + DM group. Velocities in each group was less in mid segment as compared to basal segment similar findings was established by Kucharski W³ in 2004.

E’ is reduced in both HT and DM group as compared with control and was least in HT + DM group, concluding that diastolic dysfunction is present in DM group, HT group and HT + DM combination group with diastolic dysfunction most severe in combination group. Kosmala W et al³ in their study confirmed same findings.

Gul K et al⁶ proposed that both septal E’ and lateral E’ velocities were significantly lower in diabetes than in control group. E’ in mid segments reveals similar trend in all groups but overall velocities in mid segments were less than basal segments in all groups.

E’/A’ ratio was not significantly different in control group and diabetic group. E’/A’ ratio was reduced in HT group significantly and was least in combination (HT and DM) group. Similar findings were proposed by Monika Przewlocka-Kosmala M⁷ in 2004.

The present study demonstrates the absence of significant difference in LV structure and function between DM group and healthy control subjects when analyzed by conventional doppler methods as transmitral inflow pattern and DT but when analyzed with tissue doppler as MPSV and E’ there is significant impairment in LV function in DM as compared to control. 

DM is associated in most cases with HT, obesity and high prevalence of coronary artery disease and microangiopathic complication which are known to impair LV diastolic function. Nevertheless some studies have demonstrated an impairment in LV diastolic function in diabetic patients independent of hypertension and coronary artery disease, suggesting that this abnormality could be an early manifestation of a specific diabetic cardiomyopathy (Raev DC).⁸ Most of these studies have been performed in small sample sized population. While some studies reported significant abnormalities of diastolic function (Annonu AK et al)⁹ other demonstrated no differences from healthy control subjects (Romanens Met et al).¹⁰ By the way, evidence of an intrinsic diastolic dysfunction in DM remains questionable.

Tissue doppler imaging (TDI) is a new ultrasound method that records regional systolic and diastolic velocities with in the myocardium. A good correlation has been found between the initial diastolic peak velocity and ventricular relaxation measurements obtained through invasive methods (Nagueh SF et al).¹¹

Previous studies have shown that the early diastolic velocity recorded at the mitral annulus is reduced in patients with relaxation abnormality and is sensitive to change not identified by conventional mitral doppler index (Severino S et al).¹² To date, only few studies have used this new doppler method in the assessment of subclinical myocardial disease in DM (Fang ZY et al).¹³

Fang ZY et al¹³ have recently reported that regional diastolic myocardial function (peak early diastolic velocity obtained at the septal annulus) is significantly reduced in a population of DM patients compared with controls.

The observation that DM was associated with abnormal LV diastolic filling suggest that hyperglycemia may contribute to the pathogenetic mechanism of abnormal ventricular relaxation in DM. Interstitial accumulation of advanced glycated end products (AGEs) which include collagen, elastin and other connective tissue proteins as well as fibrosis in the myocardium have been reported in human diabetic heart which can increase end diastolic stiffness as well as LV mass. Quantization of fibrosis in hypertensive, diabetes and hypertensive diabetic heart has revealed the lowest proportion in hypertensive hearts and the highest in hypertensive diabetic hearts with diabetic hearts in the mid range (von Hoeven KH et al).¹⁴

Another factor linking DM and abnormal LV relaxation may be the presence of coronary artery disease, although that may be the etiology in some cases. All of our subjects had negative stress ECG and normal wall motion with ejection fraction >55%.

Our study confirms and substantially extends previous findings by revealing additional cardiovascular abnormalities associated with DM. DM is associated with abnormal LV relaxation similar to the well known impaired relaxation associated with HT. Abnormal LV relaxation seen in DM independent of other factors, may contribute to the increase incidence of congestive heart failure despite normal LV ejection fraction, thereby being another cause of clinical cardiovascular morbidity. In addition, reduced mitral E/A ratio is independently associated with increased all cause mortality as well as cardiovascular morbidity (Bella JN et al).¹⁵ Abnormal LV filling pattern suggests that degree of hyperglycemia may play a role in the pathogenesis of diastolic dysfunction in DM. A prospective trial is needed to determine whether better glycemic control will improve LV structure and function in adults with DM.

Conclusion

Conventional doppler can detect diastolic dysfunction in HT and HT + DM group while in DM group conventional doppler was not able to detect diastolic dysfunction. While tissue doppler imaging was able to detect LV diastolic dysfunction in DM, HT and DM + HT group.

References

1. Kannel WB, McGee DL. Diabetes and cardiovascular disease. JAMA. 1979;241(19):2035-8.
2. Ruderman NB, Haudenschild C. Diabetes as an atherogenic factor. Prog Cardiovasc Dis. 1984;26(5):373-412.
3. Kosmala W, Kucharski W, Przewlocka-Kosmala M, Mazurek W, et al. Comparison of left ventricular function by tissue Doppler imaging in patients with diabetes mellitus without systemic hypertension versus diabetes mellitus with systemic hypertension. Am J Cardiol 2004;94(3):395-9.
4. Liu JE, Palmieri V, Roman MJ, Bella JN, Fabsitz R, Howard BV, et al. The impact of diabetes on left ventricular filling pattern in normotensive and hypertensive adults: the strong Heart Study. J Am Coll Cardiol. 2001;37(7):1943-9.
5. Shapiro LM. Echocardiographic features of impaired ventricular function in diabetes mellitus. Br Hear J. 1982;47 (5):439-44.
6. Gul K, Celebi AS, Kacmaz F, Ozcan OC, Ustun I, Berker D, et al. Tissue Doppler imaging must be performed to detect early left ventricular dysfunction in patients with type 1 diabetes mellitus. Eur J Echocardiogr. 2009;10(7):841-6.
7. Kosmala W, Kucharski W, Przewlocka-Kosmala M, Mazurek W. Comparison of left ventricular function by tissue Doppler imaging in patients with diabetes mellitus without systemic hypertension versus diabetes mellitus with systemic hypertension. Am J Cardiol. 2004;94(3):395-9.
8. Raev DC. Which left ventricular function is impaired earlier in the evolution of diabetic cardiomyopathy? An echocardiographic study of young type I diabetic patients. Diabetes Care. 1994;17(7):633-9.
9. Annonu AK, Fattah AA, Mokhtar MS, Ghareeb S, Elhendy A. Left ventricular systolic and diastolic functional abnormalities in asymptomatic patients with non-insulin-dependent diabetes mellitus. J Am Soc Echocardiogr. 2001; 14(9):885-91.
10. Romanens M, Fankhauser S, Saner B, Michaud L, Saner H. No evidence for systolic or diastolic left ventricular dysfunction at rest in selected patients with long-term type I diabetes mellitus. Eur J Heart Fail. 1999;1(2):169-75.
11. Nagueh SF, Middleton KJ, Kopelen HA, Zoghbi WA, Quinones MA. Doppler tissue imaging: a noninvasive technique for evaluation of left ventricular relaxation and estimation of filling pressures. J Am Coll Cardiol. 1997;30 (6):1527-33.
12. Severino S, Caso P, Galderisi M, Simone LD, Petrocelli A, de Divitiis O, et al. Use of pulsed Doppler tissue imaging to assess regional left ventricular diastolic dysfunction in hypertrophic cardiomyopathy. Am J Cardiol. 1998;82 (11):1394-98.
13. Fang ZY, Yuda S, Anderson V, Short L, Case C, Marwick TH. Echocardiographic detection of early diabetic myocardial disease. J Am Coll Cardiol. 2003;41(4):611-7.
14. van Hoeven KH, Factor SM. A comparison of the pathological spectrum of hypertensive, diabetic, and hypertensive-diabetic heart disease. Circulation. 1990;82(3): 848-55.
15. Bella JN, Palmieri V, Roman MJ, et al. Prognostic significance of abnormal peak early to late diastolic filling ratio in middle-aged to elderly American Indians: the Strong Heart Study (abstr). J Am Coll Cardiol 2000;35:293A.